Загрузка видео...

Не удалось загрузить видео

На главную

Many 3D generators output Gaussian Splats (3DGS) for fast rendering, flexible deployment, and high visual fidelity. Static 3DGS aren't world models (no dynamics/semantics) but a true world model must allow distilling 3D-consistent representations for any given time step (3DGS/meshes). This post-distillation serves a dual purpose: 1) validates physical consistency...

MattNiessner's profile picture

Matthias Niessner

47,629 subscribers

26,226 просмотров • 3 месяцев назад •via X (Twitter)

LIVE

Anya Rossi• Live Now

Private livecam show

Комментарии: 0

Нет доступных комментариев

Здесь появятся комментарии из оригинального поста

Похожие видео

Beyond photos/videos, what's the next medium for sharing visual experiences? 🤔 NeRF/3DGS are great but only capture a static 3D world and ignore ambient scene dynamics. In this paper, we show how we can reconstruct/render these essential scene elements even with a single causal video.
0:10

Beyond photos/videos, what's the next medium for sharing visual experiences? 🤔 NeRF/3DGS are great but only capture a static 3D world and ignore ambient scene dynamics. In this paper, we show how we can reconstruct/render these essential scene elements even with a single causal video.

Jia-Bin Huang

28,755 просмотров • 2 лет назад

🧊 Splats → Voxels: Bridging 3D Gaussian Splatting with Dynamic Voxels World Labs Just added this last tool to the GS editor plug-in that can convert any 3DGS splats into interactive voxel worlds with full physics support! 🚀 3 Rendering Modes: 🔵 CPU Mode - GameObjects for each voxel. Max physics interaction, perfect for destructible environments 🟢 GPU Mode - DrawMeshInstancedIndirect. Renders 100K+ voxels at 90fps. Pure visual, no physics overhead 🟡 Hybrid Mode - Best of both! GPU renders distant voxels Building this so you can literally punch through a GS World and watch it crumble 🥊 #GaussianSplatting #unity3d #GameDev #VR #3DGS #Voxels #Physics #madewithunity
0:11

🧊 Splats → Voxels: Bridging 3D Gaussian Splatting with Dynamic Voxels World Labs Just added this last tool to the GS editor plug-in that can convert any 3DGS splats into interactive voxel worlds with full physics support! 🚀 3 Rendering Modes: 🔵 CPU Mode - GameObjects for each voxel. Max physics interaction, perfect for destructible environments 🟢 GPU Mode - DrawMeshInstancedIndirect. Renders 100K+ voxels at 90fps. Pure visual, no physics overhead 🟡 Hybrid Mode - Best of both! GPU renders distant voxels Building this so you can literally punch through a GS World and watch it crumble 🥊 #GaussianSplatting #unity3d #GameDev #VR #3DGS #Voxels #Physics #madewithunity

Daniel Skaale

28,534 просмотров • 5 месяцев назад

EDGS: Eliminating Densification for Efficient Convergence of 3DGS Contributions: • We show that initial triangulation based on 2D correspondences can replace the incremental refinement process, fundamentally changing how 3DGS models allocate resources. • Our method reduces the path each Gaussian must travel in parameter space. Careful initialization not only accelerates convergence but also guides optimization toward a convergence point corresponding to lower reconstruction error and thus higher reconstruction quality. • Our approach outperforms both speed-optimized and quality-focused state-of-the-art models while using only half the splats of standard 3DGS. By improving initialization rather than altering the optimization process, this method is compatible with other 3DGS acceleration techniques, making it a flexible enhancement to existing models.
0:15

EDGS: Eliminating Densification for Efficient Convergence of 3DGS Contributions: • We show that initial triangulation based on 2D correspondences can replace the incremental refinement process, fundamentally changing how 3DGS models allocate resources. • Our method reduces the path each Gaussian must travel in parameter space. Careful initialization not only accelerates convergence but also guides optimization toward a convergence point corresponding to lower reconstruction error and thus higher reconstruction quality. • Our approach outperforms both speed-optimized and quality-focused state-of-the-art models while using only half the splats of standard 3DGS. By improving initialization rather than altering the optimization process, this method is compatible with other 3DGS acceleration techniques, making it a flexible enhancement to existing models.

MrNeRF

124,002 просмотров • 1 год назад

The context size of video world models is only a few frames. Like a human with severe memory loss! We design a long-term memory for world models based on explicit 3D representations inspired by the human mind. This enables long-term consistency. 1/3
0:12

The context size of video world models is only a few frames. Like a human with severe memory loss! We design a long-term memory for world models based on explicit 3D representations inspired by the human mind. This enables long-term consistency. 1/3

Gordon Wetzstein

34,796 просмотров • 1 год назад

Create a 3D model from a single image, set of images or a text prompt in < 1 minute 😮‍💨 This new AI paper called CAT3D shows us that it’ll keep getting easier to produce 3D models from 2D images — whether it’s a sparser real world 3D scan (a few photos instead of hundreds) or your favorite 2D image generator like Midjourney (just an image). How does this magic work? “This architecture is similar to video diffusion models, but with camera pose embeddings for each image instead of time embeddings. The generated views are passed into a robust 3D reconstruction pipeline to create the 3D representation (Zip-NeRF or 3DGS)”
0:18

Create a 3D model from a single image, set of images or a text prompt in < 1 minute 😮‍💨 This new AI paper called CAT3D shows us that it’ll keep getting easier to produce 3D models from 2D images — whether it’s a sparser real world 3D scan (a few photos instead of hundreds) or your favorite 2D image generator like Midjourney (just an image). How does this magic work? “This architecture is similar to video diffusion models, but with camera pose embeddings for each image instead of time embeddings. The generated views are passed into a robust 3D reconstruction pipeline to create the 3D representation (Zip-NeRF or 3DGS)”

Bilawal Sidhu

92,760 просмотров • 2 лет назад

NeuralGS Bridging Neural Fields and 3D Gaussian Splatting for Compact 3D Representations
0:10

NeuralGS Bridging Neural Fields and 3D Gaussian Splatting for Compact 3D Representations

AK

68,496 просмотров • 1 год назад

Meet 3D Gaussian Splatting - a 3D AI breakthrough that’s changing the game for real-time 3D rendering. It generates 3D world with state-of-the-art visuals at fast speeds: 100+ fps at 1080p. Here’s the magic behind it:
0:13

Meet 3D Gaussian Splatting - a 3D AI breakthrough that’s changing the game for real-time 3D rendering. It generates 3D world with state-of-the-art visuals at fast speeds: 100+ fps at 1080p. Here’s the magic behind it:

el.cine

60,115 просмотров • 1 год назад

Procedural Gaussian Splats in PlayCanvas 🔥 Everything becomes splats: Images. Text. Lines. One unified render, seamlessly blending with 3DGS models.
0:17

Procedural Gaussian Splats in PlayCanvas 🔥 Everything becomes splats: Images. Text. Lines. One unified render, seamlessly blending with 3DGS models.

Martin Valigursky

19,026 просмотров • 4 месяцев назад

Human Hair Reconstruction with Strand-Aligned 3D Gaussians Contributions (cited): – We propose a new 3D line lifting scheme that uses a modified 3DGS reconstruction technique to lift 2D orientation maps into a 3D field while also providing refinement of the camera parameters; – We introduce a dual representation of hair strand polylines and 3D Gaussians to achieve differentiable rasterization of hair strands and leverage photometric constraints for strand-based hair reconstruction; – Based on these components, we propose a coarse-to-fine optimization method for prior-guided hair reconstruction that leverages both latent and explicit representations of the hairstyle.
0:13

Human Hair Reconstruction with Strand-Aligned 3D Gaussians Contributions (cited): – We propose a new 3D line lifting scheme that uses a modified 3DGS reconstruction technique to lift 2D orientation maps into a 3D field while also providing refinement of the camera parameters; – We introduce a dual representation of hair strand polylines and 3D Gaussians to achieve differentiable rasterization of hair strands and leverage photometric constraints for strand-based hair reconstruction; – Based on these components, we propose a coarse-to-fine optimization method for prior-guided hair reconstruction that leverages both latent and explicit representations of the hairstyle.

MrNeRF

106,497 просмотров • 1 год назад

(1/2) Check out "𝐏𝐨𝐥𝐲𝐃𝐢𝐟𝐟: Generating 3D Polygonal Meshes with Diffusion Models"! Our model operates directly on the polygons of 3D meshes and generates novel shapes as output through an iterative diffusion process.
0:11

(1/2) Check out "𝐏𝐨𝐥𝐲𝐃𝐢𝐟𝐟: Generating 3D Polygonal Meshes with Diffusion Models"! Our model operates directly on the polygons of 3D meshes and generates novel shapes as output through an iterative diffusion process.

Matthias Niessner

57,940 просмотров • 2 лет назад

📢 Our lab has been exploring 3D world models for years — and we’re thrilled to share **PhysTwin**: a milestone that reconstructs object appearance, geometry, and dynamics from just a few seconds of interaction! Led by the amazing Hanxiao Jiang 👉 PhysTwin combines **Gaussian splatting** with **inverse dynamics optimization** based on simple **spring-mass** systems. ⚙️ The result? Real-time, action-conditioned 3D video prediction under novel interactions (i.e., 3D world models). 🔑 A few key takeaways: 1. Having the right structure (e.g., particles/masses) helps navigate the trade-off between sample efficiency, generalization, and broad applicability. 2. Visual foundation models (VFMs) have matured to the point where they can provide rich supervision for world modeling (e.g., tracking, shape completion). 3. Beyond VFMs, many crucial components have come together in recent years: Gaussian splats for rendering, NVIDIA Warp for high-performance simulation, and scene/asset generation from a wide range of labs and companies. The future of 3D world models is looking bright! ✨ 4. The resulting digital twin supports a wide range of downstream applications—especially in data generation and policy evaluation, thanks to its realistic rendering and simulation capabilities. 🎥 All code and data to reproduce the results, along with interactive demos, are available on the website. Check the following visualizations of: (1) observations, (2) reconstructed state/actions, (3) interactive digital twins, and (4) the overlays between real-world robot teleoperation and our model’s open-loop predictions.
0:05

📢 Our lab has been exploring 3D world models for years — and we’re thrilled to share **PhysTwin**: a milestone that reconstructs object appearance, geometry, and dynamics from just a few seconds of interaction! Led by the amazing Hanxiao Jiang 👉 PhysTwin combines **Gaussian splatting** with **inverse dynamics optimization** based on simple **spring-mass** systems. ⚙️ The result? Real-time, action-conditioned 3D video prediction under novel interactions (i.e., 3D world models). 🔑 A few key takeaways: 1. Having the right structure (e.g., particles/masses) helps navigate the trade-off between sample efficiency, generalization, and broad applicability. 2. Visual foundation models (VFMs) have matured to the point where they can provide rich supervision for world modeling (e.g., tracking, shape completion). 3. Beyond VFMs, many crucial components have come together in recent years: Gaussian splats for rendering, NVIDIA Warp for high-performance simulation, and scene/asset generation from a wide range of labs and companies. The future of 3D world models is looking bright! ✨ 4. The resulting digital twin supports a wide range of downstream applications—especially in data generation and policy evaluation, thanks to its realistic rendering and simulation capabilities. 🎥 All code and data to reproduce the results, along with interactive demos, are available on the website. Check the following visualizations of: (1) observations, (2) reconstructed state/actions, (3) interactive digital twins, and (4) the overlays between real-world robot teleoperation and our model’s open-loop predictions.

Yunzhu Li

25,279 просмотров • 1 год назад

🔥Happy New Year all! 🧨 In 2026 Splats will be like pixels! Ultra dense 3D Gaussian Splatting VFX Backdrops! In 360° → 3DGS pipeline >8M splats per scene Zero compromises on detail! Every surface. Every edge. Every reflection. This is what photorealism should look like:-) #GaussianSplatting #unity3d #GameDev #VR #3DGS #Physics #madewithunity
0:16

🔥Happy New Year all! 🧨 In 2026 Splats will be like pixels! Ultra dense 3D Gaussian Splatting VFX Backdrops! In 360° → 3DGS pipeline >8M splats per scene Zero compromises on detail! Every surface. Every edge. Every reflection. This is what photorealism should look like:-) #GaussianSplatting #unity3d #GameDev #VR #3DGS #Physics #madewithunity

Daniel Skaale

16,806 просмотров • 5 месяцев назад

F3D-Gaus: Feed-forward 3D-aware Generation on ImageNet with Cycle-Consistent Gaussian Splatting Contributions: • We pioneer 3D-aware generation using generalizable feed-forward Gaussian Splatting representation, achieving significant efficiency and favorable rendering quality on monocular datasets. • We significantly advance the capability of pixel-aligned Gaussian Splatting representations by designing a self-supervised cycle training strategy specifically tailored for monocular datasets. • We further mitigate the artifacts of 3D-aware representations caused by large viewpoint shifts by introducing geometry-aware video priors.
0:16

F3D-Gaus: Feed-forward 3D-aware Generation on ImageNet with Cycle-Consistent Gaussian Splatting Contributions: • We pioneer 3D-aware generation using generalizable feed-forward Gaussian Splatting representation, achieving significant efficiency and favorable rendering quality on monocular datasets. • We significantly advance the capability of pixel-aligned Gaussian Splatting representations by designing a self-supervised cycle training strategy specifically tailored for monocular datasets. • We further mitigate the artifacts of 3D-aware representations caused by large viewpoint shifts by introducing geometry-aware video priors.

MrNeRF

14,229 просмотров • 1 год назад

A blueberry. 90 stacks, 68 photos each. 1.48M splats. #3dgs
0:12

A blueberry. 90 stacks, 68 photos each. 1.48M splats. #3dgs

Dany Bittel

720,688 просмотров • 1 месяц назад

Wonderland: Navigating 3D Scenes from a Single Image Contributions: • First, we introduce a representation for controllable 3D generation by leveraging the generative priors from camera-guided video diffusion models. Unlike image models, video diffusion models are trained on extensive video datasets. This enables them to capture comprehensive spatial relationships within scenes across multiple views and embed a form of "3D awareness" in their latent space, which allows us to maintain 3D consistency in novel view synthesis. • Second, to achieve controllable novel view generation, we empower video models with precise control over specified camera motions. We introduce a novel dual-branch conditioning mechanism that effectively incorporates desired diverse camera trajectories into the video diffusion model. This enables expansion of a single image into a multi-view consistent capture of a 3D scene with precise pose control. • Third, to achieve efficient 3D reconstruction, we directly transform video latents into 3DGS. We propose a novel latent-based large reconstruction model (LaLRM) that lifts video latents to 3D in a feed-forward manner. With this design, during inference, our model directly predicts 3DGS from a single input image, effectively aligning the generation and reconstruction tasks—and bridging image space and 3D space—through the video latent space. Compared with reconstructing scenes from images, the video latent space offers a 256× spatial-temporal reduction while retaining essential and consistent 3D structural details. Such a high degree of compression is crucial, as it allows the LaLRM to handle a wider range of 3D scenes within the reconstruction framework, with the same memory constraints.
0:12

Wonderland: Navigating 3D Scenes from a Single Image Contributions: • First, we introduce a representation for controllable 3D generation by leveraging the generative priors from camera-guided video diffusion models. Unlike image models, video diffusion models are trained on extensive video datasets. This enables them to capture comprehensive spatial relationships within scenes across multiple views and embed a form of "3D awareness" in their latent space, which allows us to maintain 3D consistency in novel view synthesis. • Second, to achieve controllable novel view generation, we empower video models with precise control over specified camera motions. We introduce a novel dual-branch conditioning mechanism that effectively incorporates desired diverse camera trajectories into the video diffusion model. This enables expansion of a single image into a multi-view consistent capture of a 3D scene with precise pose control. • Third, to achieve efficient 3D reconstruction, we directly transform video latents into 3DGS. We propose a novel latent-based large reconstruction model (LaLRM) that lifts video latents to 3D in a feed-forward manner. With this design, during inference, our model directly predicts 3DGS from a single input image, effectively aligning the generation and reconstruction tasks—and bridging image space and 3D space—through the video latent space. Compared with reconstructing scenes from images, the video latent space offers a 256× spatial-temporal reduction while retaining essential and consistent 3D structural details. Such a high degree of compression is crucial, as it allows the LaLRM to handle a wider range of 3D scenes within the reconstruction framework, with the same memory constraints.

MrNeRF

52,801 просмотров • 1 год назад

2025 is year of 3D AI Rodin AI 1.5 now can not only generate 3D model with clean meshes but also create high quality textures You can export all the models to 3D software to create a new world
0:18

2025 is year of 3D AI Rodin AI 1.5 now can not only generate 3D model with clean meshes but also create high quality textures You can export all the models to 3D software to create a new world

el.cine

92,524 просмотров • 1 год назад

We just open-sourced 3DGS render blender addon v4.0 ( It has been a while, but we are still updating this 3D Gaussian Splatting tool! In this version, we added a real-time rendering mode, so say goodbye to those laggy displays in the viewport! Here is an example: I imported a file that contains one million gaussians, super smooth.
0:15

We just open-sourced 3DGS render blender addon v4.0 ( It has been a while, but we are still updating this 3D Gaussian Splatting tool! In this version, we added a real-time rendering mode, so say goodbye to those laggy displays in the viewport! Here is an example: I imported a file that contains one million gaussians, super smooth.

Chris make some 3D scans

68,522 просмотров • 8 месяцев назад

I can no longer walk around a city without seeing a machine-readable 3D model in my head. This is a geometric and semantic 3D model of San Francisco. These maps connect the world of bits and atoms, enabling visual positioning, 3D navigation and geospatial intelligence.
0:12

I can no longer walk around a city without seeing a machine-readable 3D model in my head. This is a geometric and semantic 3D model of San Francisco. These maps connect the world of bits and atoms, enabling visual positioning, 3D navigation and geospatial intelligence.

Bilawal Sidhu

260,180 просмотров • 1 год назад

A Raspberry. 90 stacks, 68 photos each. 2.37M splats. #3dgs
0:15

A Raspberry. 90 stacks, 68 photos each. 2.37M splats. #3dgs

Dany Bittel

7,128,901 просмотров • 1 месяц назад

Mujoco + 3DGS on browser, imported from Fei-Fei Li ‘s World Labs Import any robot into any 3DGS scene Play the simulation on your phone EE control on XLeRobot and Franka
0:11

Mujoco + 3DGS on browser, imported from Fei-Fei Li ‘s World Labs Import any robot into any 3DGS scene Play the simulation on your phone EE control on XLeRobot and Franka

Vector Wang

14,618 просмотров • 4 месяцев назад